Illuminating economics’ biggest error

If you are a regular reader of these missives, you will know that I hardly ever write about models and theories. This is because I think that economics and finance theory is so full of errors and holes as to make it almost useless for practical purposes. Instead, prefer to look at the world as it is, not as it should be or as economists model it.

A case in point is what I still consider to be the biggest error economists have ever made: the assumption that you can model humans as acting in their self-interest and nothing else. This assumption famously is traced back to Adam Smith’s Wealth of Nation where he wrote the famous lines:

It is not from the benevolence of the butcher, the brewer, or the baker that we expect our dinner, but from their regard to their own interest.

Ironically, Adam Smith was smart enough to know that this description of humans and self-interested beings is incomplete. He wrote an entire second book, The Theory of Moral Sentiments, about the flaws of this assumption. Yet, generations of classical economists have ignored the second book and elevated the first book to something like a sacred test. By now we have almost 250 years of experimental evidence and data that shows that humans are born to cooperate, yet in finance, there are still so many people – especially in the United States, I may say – who still are adamant that this assumption is all one needs to understand markets.

But a new experiment amongst children from Germany and Kenya has shown nicely that cooperation is a stronger force than self-interest, even at a very young age. The experimenters recruited 104 children aged five or six from a Western, individualistic society (in this case Germany), and 103 children of the same age from a very different, more collectivist society (in this case Kenya). Then they asked them to perform the famous marshmallow test. If you don’t know what the marshmallow test is, then watch this video, which is one of my all-time favourites.

The marshmallow test explained

Source: TED

The trick with this new experiment is to use the marshmallow test to see if young children would be willing to delay gratification more if they are asked to cooperate. The experimenter randomly paired up two children and then assigned one of three different trials to these two children:

Solo condition: Child 1 is given a cookie by an adult. The adult tells the child that he has to leave the room. If he comes back and the cookie is still there and the child still sits in the chair, she will get a second cookie. This is a test if a child can act in its self-interest to get a second cookie by delaying gratification.

Interdependence: Child 1 is given a cookie by an adult. Child 2 is also given a cookie by the adult. Both child 1 and child 2 are then told that they play the game together. The adult has to leave the room and if he comes back and both children have not eaten their cookies and are still in their chairs, they both get a second cookie. If one of the children eats their cookie, neither one will get a second cookie. This is a test if children can cooperate to get a second cookie by both delaying gratification. The risk for each child is that the other child will eat the cookie and they both get only one.

And the results are: If the children are asked to cooperate (Interdependence) a far larger share of children is able to resist temptation and wait until the adult comes back. The need to cooperate leads to better outcomes for both children and much better outcomes than a situation when the children are simply asked to rely on their self-interest. We humans are not born to act in our self-interest. We are born as social animals and try to act in the best interest of our group. Within our group (however this group may be defined), we cooperate intensively, while we compete with other groups. But even when it comes to group competition, cooperation will lead to better outcomes. And if politicians and economists would finally adopt that as the foundation of their ‘models’ we would probably have better societal outcomes as well. 

Share of children able to wait in self-interested condition (Solo) or cooperation condition (Interdependence)

Source: Koomen et al. (2020). Note: Green bars are German children, Orange bars are Kenyan children.